AHAP: Reconstructing Arbitrary Humans from Arbitrary Perspectives with Geometric Priors

TL;DR

AHAP is a fast, calibration-free framework for reconstructing 3D humans from multiple views using geometric priors, outperforming traditional methods in speed and accuracy.

Contribution

It introduces a novel, feed-forward approach that combines multi-view geometry and learnable association modules for arbitrary human reconstruction without camera calibration.

Findings

Achieves competitive 3D human reconstruction accuracy.

Significantly faster (180x) than optimization-based methods.

Effective multi-view human localization and identity association.

Abstract

Reconstructing 3D humans from images captured at multiple perspectives typically requires pre-calibration, like using checkerboards or MVS algorithms, which limits scalability and applicability in diverse real-world scenarios. In this work, we present AHAP (Reconstructing Arbitrary Humans from Arbitrary Perspectives), a feed-forward framework for reconstructing arbitrary humans from arbitrary camera perspectives without requiring camera calibration. Our core lies in the effective fusion of multi-view geometry to assist human association, reconstruction and localization. Specifically, we use a Cross-View Identity Association module through learnable person queries and soft assignment, supervised by contrastive learning to resolve cross-view human identity association. A Human Head fuses cross-view features and scene context for SMPL prediction, guided by cross-view reprojection losses to enforce body pose consistency. Additionally, multi-view geometry eliminates the depth ambiguity inherent in monocular methods, providing more precise 3D human localization through multi-view triangulation. Experiments on EgoHumans and EgoExo4D demonstrate that AHAP achieves competitive performance on both world-space human reconstruction and camera pose estimation, while being 180$\times$ faster than optimization-based approaches.